Background: The concept of increased medical errors in July, secondary to new residents (the "July phenomenon"), often receives considerable attention without supporting evidence. A recent study reported a 41% increase in mortality during July and August in general surgery patients. The objective of this study was to determine if a July phenomenon existed in a Level I trauma center with an attending present at all times.
Study design: A retrospective cohort study was conducted at an academic, tertiary Level I trauma center. Blunt trauma patients admitted during a 5-year period were compared for differences in outcomes by month and quarter. Chi-square and analysis of variance were used for categorical and continuous variables where appropriate. Linear regression was used to examine the effect of month on ventilator support days, ICU days, and minutes in the resuscitation room. Multivariable linear regression was used to examine the effect of month and quarter on mortality.
Results: A total of 12,525 patients were analyzed by month and 14,798 patients were analyzed by quarter. Overall, 68% were men and 32% women, with a mean age of 39.5 years. Mean Injury Severity Score was 12.4. Mean 24-hour transfusion requirement was 0.5 U. Mean emergency department Glasgow Coma Scale score was 14. Multivariable logistic regression failed to show month or quarter of the year to be an independent predictor of mortality after adjusting for age, Injury Severity Score, emergency department Glasgow Coma Scale score, and 24-hour transfusion requirement (c = 0.97). Linear regression failed to show any monthly variation on ventilator-support days, ICU days, or minutes in the resuscitation room.
Conclusions: The July phenomenon does not exist at this Level I trauma center with in-hospital attending supervision.